The brain is shielded against potentially harmful substances by the blood-brain barrier (BBB). The microvascular barrier between blood and brain is made up of a capillary endothelial layer surrounded by a basement membrane and tightly associated accessory cells (pericytes, astrocytes). The brain capillary endothelium is much less permeable to low-molecular weight solutes than other capillary endothelia due to an apical band of tight association between the membranes of adjoining cells, referred to as tight junctions. In addition to diminished passive diffusion, brain capillary endothelia also exhibit less fluid-phase pinocytosis than other endothelial cells. Brain capillaries possess few fenestrae and few endocytic vesicles, compared to the capillaries of other organs (see Pardridge, J. Neurovirol. 5: 556-569 (1999)). There is little transit across the BBB of large, hydrophilic molecules aside from some specific proteins such as transferrin, lactoferrin and low-density lipoproteins, which are taken up by receptor-mediated endocytosis (see Pardridge, J. Neurovirol. 5: 556-569 (1999)); Tsuji and Tamai, Adv. Drug Deliv. Rev. 36: 277-290 (1999); Kusuhara and Sugiyama, Drug Discov. Today 6:150-156 (2001); Dehouck, et al. J. Cell Biol. 138: 877-889 (1997); Fillebeen, et al. J. Biol. Chem. 274: 7011-7017 (1999)).
The blood-brain barrier (BBB) also impedes access of beneficial active agents (e.g., therapeutic drugs and diagnostic agents) to central nervous system (CNS) tissues, necessitating the use of carriers for their transit. Blood-brain barrier permeability is frequently a rate-limiting factor for the penetration of drugs or peptides into the CNS (see Pardridge, J. Neurovirol. 5: 556-569 (1999); Bickel, et al., Adv. Drug Deliv. Rev. 46: 247-279 (2001)). For example, management of the neurological manifestations of lysosomal storage diseases (LSDs) is significantly impeded by the inability of therapeutic enzymes to gain access to brain cell lysosomes. LSDs are characterized by the absence or reduced activity of specific enzymes within cellular lysosomes, resulting in the accumulation of undegraded “storage material” within the intracellular lysosome, swelling and malfunction of the lysosomes, and ultimately cellular and tissue damage. Intravenous enzyme replacement therapy (ERT) is beneficial for LSDs (e.g. MPS I, MPS II). However, the BBB blocks the free transfer of many agents from blood to brain, and LSDs that present with significant neurological sequelae (e.g. MPSI, MPS III, MLD, GM1) are not expected to be as responsive to intravenous ERT. For such diseases, a method of delivering the replacement enzyme across the BBB and into the lysosomes of the affected cells would be highly desirable.
There are several ways of circumventing the BBB to enhance brain delivery of an administered active agent include direct intra-cranial injection, transient permeabilization of the BBB, and modification of the active agent to alter tissue distribution. Direct injection of an active agent into brain tissue bypasses the vasculature completely, but suffers primarily from the risk of complications (infection, tissue damage) incurred by intra-cranial injections and poor diffusion of the active agent from the site of administration. Permeabilization of the BBB entails non-specifically compromising the BBB concomitant with injection of intravenous active agent and is accomplished through loosening tight junctions by hyperosmotic shock (e.g. intravenous mannitol). High plasma osmolarity leads to dehydration of the capillary endothelium with partial collapse of tight junctions, little selectivity in the types of blood-borne substances that gain access to the brain under these conditions, and damage over the course of a life-long regimen of treatment.
The distribution of an active agent into the brain may also be increased by transcytosis, the active transport of certain proteins from the luminal space (blood-side) to the abluminal space (brain-side) of the BBB. Transcytosis pathways are distinct from other vesicular traffic within the capillary endothelial cell and transit can occur without alteration of the transported materials. Transcytosis is a cell-type specific process mediated by receptors on the BBB endothelial surface. Attachment of an active agent to a transcytosed protein (vector or carrier) is expected to increase distribution of the active substance to the brain. In transcytosis, the vector is presumed to have a dominant effect on the distribution of the joined pair. Vector proteins include antibodies directed at receptors on the brain capillary endothelium (Pardridge, J. Neurovirol. 5: 556-569 (1999)) and ligands to such receptors (Fukuta, et al., 1994, Pharm. Res. 1994; 11(12):1681-8; Broadwell, et al., Exp Neurol. 1996; 142(1):47-65)). Antibody vectors are transported through the capillary endothelium by a process of adsorptive endocytosis (non-specific, membrane-phase endocytosis) and are far less efficiently transported than actual receptor ligands, which cross the BBB by a saturable, energy-dependent mechanism (Broadwell, et al., Exp Neurol. 1996; 142(1):47-65).
Direct administration of proteins into the brain substance has not achieved significant therapeutic effect due to diffusion barriers and the limited volume of therapeutic that can be administered. Convection-assisted diffusion has been studied via catheters placed in the brain parenchyma using slow, long-term infusions (Bobo, et al., Proc. Natl. Acad. Sci. U.S.A 91, 2076-2080 (1994); Nguyen, et al. J. Neurosurg. 98, 584-590 (2003)), but no approved therapies currently use this approach for long-term therapy. In addition, the placement of intracerebral catheters is very invasive and less desirable as a clinical alternative.
Intrathecal (IT) injection, or the administration of proteins to the cerebrospinal fluid (CSF), has also been attempted but has yielded only moderate success in a few examples of delivery via the CSF [Dittrich et al., Exp. Neurol. 141:225-239 (1996); Ochs et al., Amyotroph. Lateral. Scler. Other Motor Neuron Disord. 1:201-206 (2000); Bowes et al., Brain Res. 883:178-183 (2000)]. For nerve growth factor (NGF), the administration of the factor into the ventricle of the brain, did have some beneficial effects on the brain (Koliatsos et al., Exp. Neurol. 112, 161-173 (1991), but did not show significant diffusion into the brain substance. A major challenge in this treatment has been the tendency of the factor to bind the ependymal lining of the ventricle very tightly which prevented subsequent diffusion. Currently, there are no approved products for the treatment of brain genetic disease by therapeutic administration directly to the CSF.
The challenges in treating the brain with these and other therapeutics studied in the past have suggested that the barrier to diffusion at the brain's surface, as well as the lack of diffusion and efficacy of brain treatment, were too great an obstacle to achieve adequate therapeutic effect in the brain for any disease. Prior evidence suggests that intraventricular or intrathecal enzyme therapy would not work sufficiently to be effective, and in fact, no human studies of this approach have been published in the recent past and there are no successful examples of treatment via that route. Intrathecal injection confers an advantage over other standard treatment regimens, however, in that the CSF provides superior access to the brain and meninges. The CSF covers the brain and provides large surface area contact with cortical neurons up to 6 mm below the surface, allowing for more efficient penetration of the therapeutic into the brain tissue.
Lysosomal storage disorders affecting the nervous system demonstrate unique challenges in treating these diseases with traditional therapies. There is often a large build-up of glycosaminoglycans (GAGs) in neurons and meninges of affected individuals, leading to either mild or severe forms of the disease. For example, brain disease in severe MPS I patients is characterized by developmental delay, hydrocephalus, severe mental retardation, and eventual decline and death due to disease symptoms. Mild MPS I brain is characterized by perivascular GAG storage, hydrocephalus, learning disabilities and spinal cord compression due to swelling and scarring from storage disease. In MPS I patients in which meningeal storage is affected, the meninges are obstructed, reducing CSF resorption and leading to high pressure hydrocephalus. This aberrant lysosomal storage also leads to thickening and scarring of the meninges from storage disease.
In the lysosomal storage disorder, Gaucher disease, patients with the severe fat of the disease (type 2 and type 3) have brain disease and intravenous enzyme therapy is insufficient to effectively and adequately treat the brain. Intrathecal and intraparenchymal enzyme therapy with glucocerebrosidase, the enzyme deficient in Gaucher disease, has succeeded in getting into the brain but did not successfully treat the brain storage (Zirzow et al., Neurochem. Res. 24:301-305. 1999). At this time, no brain disease resulting from a lysosomal disorder has successfully been treated by any means available.
Thus, there remains a need in the art to develop methods which effectively treat lysosomal storage disorders through effective administration of enzyme replacement therapy. More particularly, a need exists for more effective methods of administration of compounds and compositions that can more efficiently deliver active agents to the brain and central nervous system for the treatment of lysosomal storage disorders.